Covid-19 Impact on Bicycle Usage

Data Systems

The discussion on API and data systems highlights the disparity between New York and Colorado in terms of bicycle data collection. Abdullah Kurkcu explains that New York offers a sophisticated API for real-time data access, while Colorado relies on continuous counters that provide historical data dating back to 2014 1. This longitudinal data is invaluable for analyzing trends over time, especially pre-pandemic. Kurkcu shares his personal journey into transportation, sparked by his experiences with traffic congestion in Turkey, emphasizing the importance of data-driven solutions 2.

There's always a better way to move people and goods from one location to the other. And that should be backed up by data, of course.

--- Abdullah Kurkcu

The availability of mobile devices has further enriched data collection, offering insights into travel behavior and decision-making.

   

Data Techniques

Data processing techniques in bicycle research face unique challenges, particularly when integrating socio-demographic data. Abdullah Kurkcu discusses the use of continuous count locations and socio-demographic buffers to analyze cycling behavior in the Denver area 3. However, he acknowledges the inherent biases in bicycle research due to the lack of comprehensive data and the difficulty in obtaining representative samples 4. Despite these challenges, Kurkcu emphasizes the importance of addressing collinearity in data to draw accurate conclusions.

It's really hard to get that representative sample of cyclists to be able to draw the right conclusions.

--- Abdullah Kurkcu

The integration of socio-demographic data helps in understanding the underlying reasons for cycling, although it doesn't perfectly align with ridership patterns.